SG
thesgadyal@protonmail.comPushing the boundaries of software and hardware. Former ML engineer, now diving deep into the world of high-performance computing and custom chip design.
Currently:
- Optimizing distributed training algorithms for large-scale ML models.
- Exploring the intersection of ML compilers and GPU architecture.
- Designing a custom ASIC for accelerating specific ML workloads.
- Writing about the future of AI hardware and software co-design.
Articles:
> The Future of AI Chips: Beyond Moore's Law> Scaling Mountains: Challenges in Distributed ML
> Compiler Magic: Advanced Optimization Techniques for ML
> Number Crunch: Understanding Gradient Descent in ML